Journal - The Institution of Engineers, Malaysia (Vol. 65, No. 3/4, Septtember/December 2004)
There are two main chemical polymer recycling routes namely thermal degradation, which produces a broad product range and requires high operating temperatures, and catalytic degradation, which may offer an alternative by controlling product distribution and reducing reaction temperatures. The catalytic degradation of polyolefins, which typically 60 – 70 % of polymer waste is in municipal waste streams [3], have been studied using predominantly pure catalysts, namely SiO2- Al2O3[4], zeolites (ZSM-5, US-Y) and mesoporous materials (such as MCM-41) [5-7]. Strategies adopted predominantly target the formation of a stable hydrocarbon product for down- stream processing. Research is scarce on FCC catalysts with only two reports on the use of an equilibrium catalyst (E-Cat).
One is to degrade polyethylene and polystyrene at 450oC in a fluidised-bed reactor. A roughly equal mix of gas and oil product was obtained; however, no catalyst detail was mentioned [8]. The other is to degrade polypropylene at 380 oC
POLYMER WASTE RECYCLING OVER FRESH, STEAM DEACTIVATED AND USED FCC CATALYSTS
INTRODUCTION
The main challenge with the growing consumption of plastics is to keep pace with waste and waste recovery. The most common method of treating solid wastes is by disposing to landfill. This option is being curtailed under new EU Landfill Directives where 2020 UK levels must be 35% less than in 1995 [1]. In addition, Alam Flora Sdn. Bhd., a single purpose company formed by the HICOM-led Consortium who handles the solid waste management for Central and Eastern Regions of Malaysia, expects that solid waste generated in Malaysia would be doubly-increased in 20 years from 3.2 to 7.7 million tonnes per year, which give a big challenge since most of the 20 landfills have reached to their critical level [2].
APME reports [3] that feedstock recycling has great potential to enhance plastics waste recovery levels.
Unfortunately, in 1999, only 364 thousand tonnes of wastes were treated by this method and little change on 1997 figures (Table 1). Also, this process is not quite known in Malaysia. By July 2004, only 77 tonnes of plastics have been recycled from the total recycling percentage by type of material in Malaysia, which mainly by mechanical recycling (Figure 1).
Table 1: Total Plastics Consumption and Waste Recovery, Western Europe (x 1000 tonnes) [1]
ABSTRACT
Feedstock recycling of high-density polyethylene (HDPE) over fluid catalytic cracking (FCC) catalysts (6:1 ratio) was carried out using a laboratory fluidised bed reactor operating at 450oC, atmospheric pressure and nitrogen environment. Fresh and steam deactivated commercial FCC catalysts with different levels of rare earth oxide (REO) were compared as well as used FCC catalysts (E-Cats) with different levels of metal poisoning. The initial activity of the catalysts was also estimated by hexane cracking in a fixed bed reactor at 450oC. The results of HDPE degradation, in terms of yield of volatile hydrocarbon product, were: Fresh FCC catalysts > steamed FCC catalysts ≈used FCC catalysts. Overall, the product yields from HDPE cracking showed that the level of metal contamination (nickel and vanadium) did not affect the product stream generated from polymer cracking. The promising results of this study help to establish a more economic catalytic alternative for polymer waste recycling.
Keywords : polymer recycling, feedstock recycling, FCC catalysts, E-Cats Salmiaton Ali1, 2, Arthur A. Garforth2
1Department of Chemical & Environmental Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 UPM, Serdang, Selangor
2Environmental Technology Centre, School of
Chemical Engineering and Analytical Science (originally Department of Chemical Engineering, UMIST), University of Manchester, P.O. Box 88, Manchester M60 1QD, UK
1993 1995 1997 1999
Total Plastics Consumption 24 600 26 100 29 000 33 600 Total Plastics Waste 15 651 17 505 16 975 19 166 Total Plastics Waste Recovered 3 340 4 019 4 364 6 113
Mechanical Recycling 915 1 222 1 455 1 800
Feedstock Recycling 0 99 334 364
Energy Recovery 2 425 2 698 2 575 3 949
% Total Plastics Waste Recovered 21 26 26 32
Figure 1: Recycling Percentage by Type of Material in Malaysia by July 2004 [2]
in a semi-batch reactor. However, the product yield was poor compared to pure catalysts such as silica alumina, but the authors commented that increased amounts could be used as the cost was very low [9].
At UMIST, research has focused on commercial catalysts (as well as zeolites) [6, 10-13]. This paper deals with the conversion of polymer using fresh and steam deactivated catalysts as well as E-Cats with differing Ni and V levels, helping to refine costings of catalytic polymer recycling [11] for comparison with current technology [14].
MATERIALS AND METHODS
A. N-HEXANE CRACKING
Catalysts (0.1 g, 125 - 180 µm pellets) were activated at 500oC in a fixed bed quartz glass reactor by heating (1oC/min) in air (50 ml/min). The reactor was then stabilized at 450oC in flowing nitrogen (30 ml/min) and n-hexane (Aldrich, 99.9 %) was passed over the catalyst. The products were analysed by on- line gas chromatography (VARIAN 3400) [10]. A schematic-line diagram of the n-hexane cracking rig is showed in Figure 2.
B. POLYMER CRACKING
Pure HDPE (ex. BASF) with an average molecular weight of 75 000 (density 960 kg/m3) was used. The catalysts employed in this study are described in Table 2. Polymer-cracking
experiments were carried out in a fluidised bed reactor under nitrogen at 450ºC/min with a catalyst to polymer (C/P) ratio of 6:1. A schematic-line diagram of the polymer cracking rig is showed in Figure 3, whereas a detailed operation procedure and product analysis has been reported previously [10, 12].
RESULTS AND DISCUSSION
Hexane cracking and polymer degradation proceed by a carbocation mechanism, where the initially formed ions experience chain reactions via isomerisation and β-scission (monomolecular processes) and hydrogen transfer and oligomerisation (bimolecular processes) [15].
A. N-HEXANE CRACKING
Reactions of paraffins on zeolites are generally treated using simple first-order kinetic expressions [16]. However, cracking catalyst deactivation occurs at such a rapid rate that, in quantifying the rates of cracking, the kinetics of conversion must be treated simultaneously with deactivation kinetics. By following the conversion (χ) of n-hexane over the catalyst as a function of time using a time-on-stream theory of decay from a general rate of reaction term [17], an estimate of initial catalyst activity (ln t= 0) can be made (a plot of ln(-ln(1-χ)) vs.
ln tis a straight line).
The deactivation of a constant weight of catalyst was studied for a fixed flow of n-hexane in nitrogen (W/F = 27 g.h/mol, at 450oC) for up to 120 minutes. The total number of acid sites (Bronsted and Lewis), initial activities and product distributions are detailed in Table 3.
As expected, fresh FCC catalysts (Cat-1 and Cat-7) were the most active, cracking between 3 – 5% of the n-hexane feed. FCC catalysts are regenerated with steam in the FCC process and consequently lose some framework aluminium, creating defects and resulting in a lowering of acidity [18]. The presence of rare earth oxide (REO) hinders this dealumination and hence, the catalyst activity is better maintained. Indication of this stabilisation was shown in the results of hexane and polymer cracking over Cat-7S compared to Cat-1S. Cat-1S, E-
SALMIATON ALI, et al.
V1
V2
V3
V4
TC1
TC2
TC
N2
Cylinder
wt% Å m2/g ppm
Catalyst Commercial Name REO UCS MSA ZSA Ni V
Cat-1 Fresh Commercial FCC Catalyst 0.8 24.4 112 264 - - Cat-7 Fresh Commercial FCC Catalyst 9.6 24.7 90 331 - - Cat-1S *Steam Deactivated FCC Catalyst 0.8 24.3 90 198 - - Cat-7S *Steam Deactivated FCC Catalyst 9.6 24.5 72 241 - - E-Cat 1 Equilibrium FCC Catalysts 1.3 24.3 76 99 171 217 E-Cat 2 Equilibrium FCC Catalysts 1.6 24.3 32 95 5400 6580
*Steaming Conditions: 4 h/ 788ºC/ 100% Steam
1 2
3
7
8
9 10
11 12 13
6 5 4
Table 2 : Catalyst Details (Supplied by Engelhard Corporation, USA)
(1) Dry Air, (2) Dry Nitrogen, (3) Flow Controller, (4) Mass Flow Controller, (5) Flow Meter, (6) 4-tap Bubbler (containing n-hexane
at 0oC), (7) Flow Check (8) Furnace, (9) Catalyst Reactor, (10) Blank Reactor, (11) 16-loop Box, (12) GC and (13) To Vent Figure 2: Schematic Line-Diagram of the n-Hexane Cracking Rig
Polymer Feeder, 3-Zone Furnace, Sintered Distributor, Fluidising Catalyst, Fluidised Bed Reactor, Flow meter, 16-loop Automated Sample Valve in Heated Box, Gas Bag,
Gas Chromatograph, 3-Zone Digital Controller Figure 3: Schematic Diagram of Polymer Cracking Rig
Journal - The Institution of Engineers, Malaysia (Vol. 65, No. 3/4, Septtember/December 2004)
Cat1 and E-Cat2 showed similar low activities with increased olefin yield (primary cracking products) as catalytic activity reduced.
B. POLYMER CRACKING
At 450oC, the products from polymer cracking were mostly gases in the range C1- C9, coke and unreacted polymer. As expected, trends in polymer cracking (Table 4) were comparable with those of n-hexane cracking and reflected the different nature of catalysts under study.
For simplicity, catalytic polymer degradation products (P) were grouped together as hydrocarbon gases (< C5), gasoline (C5~ C9), aromatic hydrocarbon (BTX), liquids (trapped inside the side tube and the reactor) and carbonaceous deposits (trapped on the catalyst – TGA). Polymer Reacted (PR) was the weight difference between polymer feed and unreacted polymer (on the wall of the reactor and on the catalyst (TGA)).
The term "yield" was used to examine the mass balance of the overall pyrolysis, and was defined by the relationship as follows:
Yield (wt %) = P (g) ↔ 100 (1) PR (g)
Based on 0.1 g of polymer feed, fresh commercial FCC catalysts converted about 85 – 90 % of their feeds to gaseous, liquid and carbonaceous products. On the other hand, the less active steamed and equilibrium catalysts showed only 60 – 70
% conversion to the volatile products (Figure 4).
Cat-1 Cat-7 Cat-1S Cat-7S E-Cat1 E-Cat2 Bronsted + Lewis, µmol/g 606 637 116 254 28 18 Conversion, wt% 5.04 3.07 0.13 0.88 0.15 0.22
Product Distributions (moles per 100 moles converted)
C1-C4 182 189 148 159 172 146
C5-C9 12 9 30 24 19 31
BTX 0 0 0 0 0 0
Paraffins 88 84 53 64 53 36
Olefins 101 108 124 114 139 141
US-Y* Cat-1 Cat-7 Cat-1S Cat-7S E-Cat1 E-Cat2
Gaseous 55.9 75.0 71.8 50.4 55.8 64.5 65.8
Liquid 0.5 9.0 6.8 7.2 7.8 1.4 1.4
Coke 4.5 6.5 7.2 3.0 4.9 1.5 1.2
Involatile 39.1 9.5 14.2 39.4 31.5 32.6 31.6
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Gaseous Product Distribution
C1-C4 36.6 44.4 47.4 38.4 44.4 35.2 37.1
C5-C9 60.2 52.2 48.8 60.2 52.8 63.4 62.6
BTX 3.2 3.4 3.8 1.4 2.8 1.4 0.3
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Total Gaseous Product
Paraffins 48.8 53.7 60.0 31.4 48.7 23.6 23.0
Olefins 47.8 42.5 35.7 67.1 48.6 74.9 76.6
*Results for US-Y reported earlier [11]
Table 3: Initial Conversion and Product Distribution for n-Hexane Cracking at Time-on-Stream (TOS) of 1 min
Table 4: Weight % of Product Distributions at T = 450 oC; C/P = 6:1
Figure 4: Total Product Yield (wt%) at T = 450 oC; C/P = 6:1
Figure 5: Selected Olefin Products (wt%) at T = 450 oC; C/P = 6:1
Figure 6: Selected Paraffin Products (wt%) at T = 450 oC;
C/P = 6:1
Figures 5 and 6 show selected olefin and paraffin products, respectively, for the catalysts. A major active component in FCC catalysts is the zeolite US-Y and hence, product yields of pure zeolite and FCC catalysts were compared.
Again, the level of activity of the various catalysts was reflected in the amount of primary (olefin) versus secondary
SALMIATON ALI, et al.
(paraffin) products observed with the high acidities of both fresh catalysts (Cat-1 and Cat-7, Table 2) producing a high yield of secondary products. The E-Cats showed negligible loss in overall conversion of HDPE due to metal contamination and produced an olefin-rich degradation products compared with steam deactivated Cat-1S and Cat-7S. Evidence of high REO stabilization of steam deactivated catalyst, Cat-7S, was noted with a yield of balanced primary and secondary products (Figures 5 and 6).
The ratio of iso-butane to iso-butene (Figure 7) indicated the amount of hydrogen transfer that had occurred during the cracking process. The result illustrated that more bimolecular hydrogen transfer occurred in US-Y, fresh FCC catalysts and Cat-7S. Reduced acidity and deactivation of the Cat-1S and E- Cat1 and E-Cat2 showing much reduced hydrogen transfer.
CONCLUSION
Results reported here demonstrate that waste FCC catalysts, namely E-Cats [18], can catalyse the degradation of polyolefins and suggest improvements to current technology, such as the BP process in Grangemouth [14] using fluidised sand to thermally crack polymer waste at 500 oC.
Although E-Cats contaminated with heavy metals, such as nickel and vanadium, cause unwanted by-products in dehydrogenation reactions and destroy the catalyst components in FCC [18], and very less active as reported from n-hexane cracking, their use in catalytic recycling of waste polymers has been demonstrated. E-Cats would lower the operating temperature, improve the yield and selectivity of volatile products, produce potentially more valuable olefinic feedstocks than a thermal process alone and would lead to a more economic polymer waste recycling process.
Consequently, catalytic recycling of plastic wastes has the potential of contributing valuable chemical feedstock to petrochemical industries and refineries. Collection, sorting and transportation costs are the major barriers to the economic large scale processing of municipal polymer waste.
ACKNOWLEDGEMENTS
This work was performed with the financial support of the Universiti Putra Malaysia. Thank you to Dr. D.H. Harris (Engelhard Corporation) for catalysts and technical advice, and also to the UMIST Centre of Microporous Materials especially Mr. R.J. Plaisted for the use of facilities.
REFERENCES
[1] DEFRA, "Waste Strategy 2000: England and Wales Part II", Department for Environment, Food and Rural Affairs, Chap. 5, pp 64-88, 2000.
[2] Alam Flora Sdn. Bhd., 2004. <http://www.alamflora.com.my/>
[3] APME, "An Analysis of Plastics Consumption and Recovery in Western Europe 2000", Association of Plastics Manufacturers in Europe, pp 1-14, 2002.
[4] G. Luo, T. Suto, S. Yasu, and K. Kato, "Catalytic degradation of high density polyethylene and polypropylene into liquid fuel in a powder-particle fluidized bed", Polymer Degradation & Stability, vol. 70, pp. 97-103, 2000.
[5] N.D. Hesse, R. Lin, E. Bonnet, J. Copper III, and R.L. White, "In situ analysis of volatiles obtained from the catalytic cracking of polyethylene", Journal of Polymer Science,
vol. 82, pp. 3118-3125, 2001.
[6] P.N. Sharatt, Y.H. Lin, A.A. Garforth, and J. Dwyer,
"Investigation of the catalytic pyrolysis of high-density polyethylene over a HZSM-5 catalyst in a laboratory fluidised- bed reactor", Industrial Engineering Chemical Resources, vol.
36, pp. 5118-5124, 1997.
[7] Y. Uemichi, K. Takuma, and A. Ayame, "Chemical recycling of poly(ethylene) by catalytic degradation into aromatic
hydrocarbons using H-Ga-silicate", Chemical Community, vol.
18, pp.1975-1976, 1998.
[8] J. Mertinkat, A. Kirsten, M. Predel, and W. Kaminsky, W.,
"Cracking catalysts used as fluidized bed material in the Hamburg pyrolysis process", Journal of Analytical and Application Pyrolysis, vol. 49, pp. 87-95, 1999.
[9] S.C. Cardona, and A. Corma, "Tertiary recycling of polypropylene by catalytic cracking in a semibatch stirred reactor: use of spent equilibrium FCC commercial catalyst", Applied Catalysis B: Environment, vol. 25, pp. 151-156, 2000.
[10] A.A. Garforth, Y.H. Lin, P.N. Sharatt, and J. Dwyer, "Production of hydrocarbons by catalytic degradation of high density polyethylene in a laboratory fluidised-bed reactor", Applied Catalysis A: General, vol. 169,
pp. 331-342, 1998.
[11] S. Ali, A.A. Garforth, D.H. Harris, D.J. Rawlence, and Y.
Uemichi, "Polymer waste recycling over "used" catalysts", Catalysis Today, vol. 75, pp. 247-255, 2002.
[12] Y.H. Lin, P.N. Sharatt, A.A. Garforth, and J. Dwyer,
"Investigation of the deactivation of ultrastable Y zeolite by coke formation during the catalytic degradation of high density polyethylene", Thermochimica Acta, vol. 294, pp. 45-50, 1997.
[13] S. Ali, A.A. Garforth, D.H. Harris, and R.A. Shigeshi, "Interfacial Applications in Environmental Engineering", Surfactant Science Series, New York: Marcel Dekker,
vol. 108, pp. 295, 2003.
[14] Dent, and S. Hardman, "Plastics waste recycling- an alternative approach", IChemE Environmental Protection, Issue 44, pp. 1- 8, September, 1996.
[15] J. Dwyer, and D.J. Rawlence, "Fluid catalytic cracking chemistry", Catalysis Today, vol. 18, pp. 487-496, 1993.
[16] Corma, J. Planelles, J. Sanchez-Marin, and F. Tomas, "The role of different types of acid sites in the cracking of alkanes on zeolite catalysts", Journal of Catalysis, vol. 93, no. 1, pp. 30-37, 1985.
[17] R.H. Perry, and D.W. Green, Perry’s Chemical Engineers’
Handbook, Australia: McGraw Hill, 7th ed., sec. 2, pp. 99-200, 1997.
[18] J.S. Magee, and M.M. Mitchell Jr., "Fluid Catalytic Cracking:
Science and Technology", Studies Surfactant of Science &
Catalysis, The Netherlands: Elsevier Science Publishers, vol.75, pp.105, 1993.
Figure 7: Hydrogen Transfer Indication: iC4/iC4=